On Sun, Jun 26, 2016 at 12:44 PM, Jonathan Cameron <jic23@xxxxxxxxxx> wrote: > On 22/06/16 09:25, Linus Walleij wrote: >> - If no interrupt type is indicated from the DT/ACPI, >> choose IRQF_TRIGGER_HIGH so the above goes into action. > > This element is the tricky one. Previous default was (incorrectly) > IRQF_RISING so any boards relying on that will be broken. What happens is that it will request the interrupt controller (often a GPIO with interrupt capabilities) to supply an active-high level interrupt. The cases where that doesn't work is when the interrupt controller (i.e. often the GPIO driver) returns negative -EINVAL or similar from its .set_type() callback. > The problem here will be ancient boards with either wrong > device tree bindings or none at all. See stargate2.c This appears to be using pxa27x-gpio, drivers/gpio/gpio-pxa.c which has the same limitation as the Nomadik controller: only edges. So that explains why you saw the same annoying problem as me. > Clearly I can fix that one, but we still may get nasty > surprises elsewhere. Also fixing that will mean adding > platform data which we haven't previously had... > > Is it that painful to stick to a rising default? Let's do that then, surely it will be the best compromise right now. Sending v9 soon-ish. Yours, Linus Walleij -- To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
